Validate against bristol data

[Robinlovelace]

• [x] Download population and route network data for Bristol: https://github.com/ATFutures/who-data/tree/master/bristol/popdens
• [x] Assign populations to raster cells then to nodes - see here https://github.com/ATFutures/who-data/tree/master/bristol/osm
• [x] Add OD data in form of lines
• [ ] Convert lines data to OD matrix (should be a function in stplanr)
• [ ] Run the model for bristol and visualise results
• [ ] Compare with route network data here qualitatively
• [ ] Parameterize model

[mpadge]

This commit implements the working calibration code, with a final result of k = 2.9km - impressively close to my naive guess of k = 2km. That can now be considered all calibrated and done. This ticks off the above point 4, 5, and 7 (except visualisation in point 5 yet to be done).

[Robinlovelace]

Phenomenal! Great work.